Gespeichert in:
Bibliographische Detailangaben
Hauptverfasser: Sanghvi, Samyak, Ranjan, Nishant, Karmakar, Tarak
Format: Preprint
Veröffentlicht: 2025
Schlagworte:
Online-Zugang:https://arxiv.org/abs/2511.09529
Tags: Tag hinzufügen
Keine Tags, Fügen Sie den ersten Tag hinzu!
_version_ 1866910005255471104
author Sanghvi, Samyak
Ranjan, Nishant
Karmakar, Tarak
author_facet Sanghvi, Samyak
Ranjan, Nishant
Karmakar, Tarak
contents Structure-based drug design (SBDD) faces a fundamental scaling fidelity dilemma: rich pocket-aware conditioning captures interaction geometry but can be costly, often scales quadratically ($O(L^2)$) or worse with protein length ($L$), while efficient sequence-only conditioning can miss key interaction structure. We propose SiDGen, a structure-informed discrete diffusion framework that resolves this trade-off through a Topological Information Bottleneck (TIB). SiDGen leverages a learned, soft assignment mechanism to compress residue-level protein representations into a compact bottleneck enabling downstream pairwise computations on the coarse grid ($O(L^2/s^2)$). This design reduces memory and computational cost without compromising generative accuracy. Our approach achieves state-of-the-art performance on CrossDocked2020 and DUD-E benchmarks while significantly reducing pairwise-tensor memory. SiDGen bridges the gap between sequence-based efficiency and pocket-aware conditioning, offering a scalable path for high-throughput structure-based discovery.
format Preprint
id arxiv_https___arxiv_org_abs_2511_09529
institution arXiv
publishDate 2025
record_format arxiv
spellingShingle SiDGen: Structure-informed Diffusion for Generative modeling of Ligands for Proteins
Sanghvi, Samyak
Ranjan, Nishant
Karmakar, Tarak
Machine Learning
Structure-based drug design (SBDD) faces a fundamental scaling fidelity dilemma: rich pocket-aware conditioning captures interaction geometry but can be costly, often scales quadratically ($O(L^2)$) or worse with protein length ($L$), while efficient sequence-only conditioning can miss key interaction structure. We propose SiDGen, a structure-informed discrete diffusion framework that resolves this trade-off through a Topological Information Bottleneck (TIB). SiDGen leverages a learned, soft assignment mechanism to compress residue-level protein representations into a compact bottleneck enabling downstream pairwise computations on the coarse grid ($O(L^2/s^2)$). This design reduces memory and computational cost without compromising generative accuracy. Our approach achieves state-of-the-art performance on CrossDocked2020 and DUD-E benchmarks while significantly reducing pairwise-tensor memory. SiDGen bridges the gap between sequence-based efficiency and pocket-aware conditioning, offering a scalable path for high-throughput structure-based discovery.
title SiDGen: Structure-informed Diffusion for Generative modeling of Ligands for Proteins
topic Machine Learning
url https://arxiv.org/abs/2511.09529